Fingerprint Recognition on Various Authentication Sensors

Fingerprint is a very popular and an ancient biometric technology to uniquely identify a person. In this paper, a fingerprint matcher is proposed which uses the global and local adaptive binarization and global minutia features. The fingerprint data is collected using three different authentication devices based on optical sensors. The experimental results are compared with the National Institute of Standards and Technology (NIST) Bozorth algorithm and various authentication fingerprint sensors. The accuracy of the proposed algorithm has been improved significantly compared with that of the NIST Bozorth algorithm.     

指纹识别算法研究与实现

研究并实现了指纹识别过程中用到的算法,实现一套主要包括指纹图像灰度归一化和均衡化、指纹图像分割、二值化的指纹识别预处理算法、不同方向Gabor滤波的特征提取算法;尤其是采用基于Gabor滤波的指纹图像增强和纹理匹配的方法,兼顾了指纹图像的细节特征和全局特征。最后,以matlab7.0为开发平台对该方法进行验证,表明在同样的特征提取算法和匹配的条件下,改进的处理算法能够更简洁而有效地避免各种伪特征且识别率高于传统算法。     

星空背景中目标识别算法研究

为了实时捕获混杂在天基背景中的运动目标,采用基于Fourier—Mellin图像配准与边缘提取相结合的差分算法。先采用Fourier—Mellin算法和相位相关算法求取图像的旋转角度及缩放和平移参数,得出配准点,再对连续采集的两帧图像配准、差分、二值化,将差分图像与后一帧边缘提取的图像进行逻辑乘,得到目标。仿真结果表明,该算法虚警率低,适合微型航天器多目标捕获的要求。     

自动指纹识别系统预处理关键技术的研究

系统阐述了自动指纹识别系统预处理的理论、算法.在比较各种预处理方法并结合指纹图像特点的基础上,研究了指纹预处理的关键技术,包括基于直方图均衡和傅里叶变换方法的指纹图像增强、局部阈值自适应二值化等,并提出了一种新的基于方向图特性和数学形态学的指纹图像分级分割算法.实验结果表明,处理效果较好.     

指纹识别预处理方法

文章结合指纹的方向特性,提出了一套可行的指纹识别预处理算法,并对前景/背景提 取部分的算法进行了改进。通过算法仿真验证,其处理结果比较理想。     

基于EM-EKF的异类传感器系统误差配准算法

期望最大算法(EM算法)是参数估计中一种很重要的方法,在处理不完全数据中有重要应用。针对异类传感器误差配准过程中的未知参数,采用EM与EKF相结合的算法进行多传感器误差配准与航迹融合。该算法通过使似然函数的期望最大化,简化了似然函数的优化估计,EKF滤波和平滑过程实现了更加准确的期望估计,迭代算法可以有效地实现异类传感器的误差配准,使估计的系统误差收敛到似然函数的局部最优解,仿真结果显示了该算法的有效性。。     

基于红外阵列传感器的人体行为识别系统研究

随着人口老龄化的到来,为了避免发生意外事故,对老人日常活动行为进行识别和监测的安全监护系统的需求不断增长.传统的基于摄像头拍摄或者穿戴式传感器的活动状态监测系统存在着隐私保护和使用不方便等不足.为此,本文设计一种基于红外阵列传感器的人体行为识别系统.该系统通过检测环境中的温度分布和变化情况识别人体行为,不需要在老人身上佩戴任何设备,尺寸小易于安装,在黑暗环境中可正常工作,且由于采集到的是低分辨率信息,不会造成隐私泄露,对比传统方案具有明显优势.从采集到的温度分布信息中提取特征并采用K最近邻(K-Nearest Neighbor,KNN)算法实现了\     

Handwriting Character Recognition Based on Conductor/Insulator-Identifiable E-Tattoo Proximity Sensors for Blinds

Easy and convenient reading for blinds is of great significance for lowering their learning, entertainment, and communication barrier, and improving their living quality. The general solution is to learn braille, far from satisfying the meet of the blinds’ daily learning and communication. Here, a new-type conductor/insulator-identifiable e-tattoo proximity sensor is developed by simply depositing the circular interdigitated electrodes on organic semiconductor. The discriminable recognition toward conductors and insulators is realized only based on a single e-tattoo device. The sensors not only enable recognition of the protruding characters like braille based on distance difference, but also enable recognition of the handwriting ink-brush and pencil graphite characters based on the unique advantage of the sensors in distinguishing conduct and insulator. These results open a new route to realize the material category recognition, provide a user-friendly way to help the visual impaired effectively reintegrate into society, and broaden the application field of proximity sensors.     

一种指纹图像预处理算法探究

本文主要针对指纹图像预处理方法进行研究,提出了基于图像局部方差的指纹分割、基于方向场的指纹二值化和基于水平滤波器与分离滤波器组合的上下文滤波方法,最后提出了基于改进OPTA算法指纹细化方法,MATLAB仿真结果表明,本文的方法克服了噪声影响,图像细化效果良好。     

基于局部特性的指纹细节点提取算法

提出了一种基于二值指纹局部特性提取细节点的算法.该算法与快速并行细化算法相比,在提取细节点准确性相当的情况下,速度大为提升.使用FVC2002指纹数据库测试表明,该算法可靠、高效.     

Incomplete Fingerprint Recognition Based on Inproved GA-PSO

Minutiae-based algorithm plays a more and more important role in fingerprint recognition. It is greatly limited by its demand for good image quality and depen- dence on fingerprint aligning, especially for incomplete fin- gerprint recognition. Improved Genetic algorithm-Particle swarm optimization （GA-PSO） algorithm is applied to deal with these problems. The proposed algorithm improves GA-PSO in two aspects, population initialization based on reference point pre-aligning and fitness function construc- tion based on fusion of minutiae-matching and orientation- matching. The experimental results on the FVC2004 show the high effectiveness and practicability of our algorithm.     

基于小波变换的指纹特征提取方法

指纹的细节特征是指纹识别的重要依据,为提高指纹识别的速度和准确度,提出了基于离散小波变换的指纹特征提取方法。通过对指纹图像的每行进行多次haar离散小波变换,对所获得的特征向量再进行离散小波变换,最终得到指纹的特征向量,有效节省运算量,增强指纹识别系统的实时性。     

基于射频指纹的挑战-应答认证协议

无线网络存在诸多信息安全问题,解决的关键机制之一是实现通信双方的高强度认证。文中设计了基于射频指纹的双向认证协议方案,新协议基于传统的挑战-应答机制,给出存在中间人攻击的经典的Needham-Schroeder认证协议的改进实例。分析显示,改进的Needham-Schroeder认证协议能抵抗有关通信双方克隆的绝大部分攻击,有助于解决密钥泄露检测公开问题,具有更高的安全级别。新协议因此具有应用价值。     

基于指纹特征数字水印算法的身份认证技术研究

本文利用指纹特征信息生成数字水印信号,并隐藏在身份证件之中.以身份证件作为指纹信息的载体,取代了传统指纹识别系统中的指纹特征参考信息数据库,提高整个自动指纹识别系统(AFIS)的安全性.同时,嵌入指纹特征信息的身份证件与用户存在本质联系,防止证件被盗用,实现对证件的保护.实验证明,使用含水印的证件在保证指纹识别系统正常工作的前提下,提高了验证的安全性,并且水印算法具有一定的鲁棒性.     

Tracking Your Browser with High-Performance Browser Fingerprint Recognition Model

As the cyber security has attracted great attention in recent years,and with all kinds of tools’(such as Network Agent,VPN and so on)help,traditional methods of tracking users like log analysis and cookie have been not that effective.Especially for some privacy sensitive users who changed their browser configuration frequently to hide themselves.The Browser Fingerprinting technology proposed by Electronic Frontier Foundation(EFF)gives a new approach of tracking users,and then our team designed an enhanced fingerprint dealing solution based on browser fingerprinting technology.Our enhanced solution plays well in recognizing the similar fingerprints,but it is not that efficient.Nowadays we improve the algorithm and propose a high-performance,efficient Browser Fingerprint Recognition Model.Our new model reforms the fingerprint items set by EFF and propose a Fingerprint Tracking Algorithm(FTA)to deal with collected data.It can associate users with some browser configuration changes in different periods of time quickly and precisely.Through testing with the experimental website built on the public network,we prove the high-performance and efficiency of our algorithm with a 20%time-consuming decrease than ever.     

基于多层次分块的指纹图像二值化

自适应局部阈值二值化算法,其效果主要取决于能否准确地估计局部阈值.鉴于在指纹奇异点附近及方向变化较快的脊线、谷线区域,局部阈值很难精确得到,提出了一种多层次分块的指纹图像局部阈值的求取方法,并把这种方法用于指纹图像的二值化.实践表明,算法工作稳定,效果良好,鲁棒性强,对低质量图像具有显著的增强效果.     

基于连续方向图的指纹图像增强算法

提出一种新的指纹图像增强算法。先得到指纹图像的连续方向图,再根据此方向图设计方向滤波模板进行滤波,最后得到指纹的二值化图像。实验结果表明,该算法简单有效,运算速度快,为指纹的特征提取和识别奠定了坚实的基础。     

一种基于方向场的指纹图像滤波算法

指纹识别是当今应用非常广泛的身份识别方法,它以指纹的唯一性和稳定性为基础,指纹识别具有方便、安全、可靠等特点,使自动指纹识别系统在人们的生活中扮演着一个十分重要的角色。对图像预处理进行了研究,图像预处理主要包括图像的滤波、二值化及细化三个步骤。滤波采用的是基于方向场的滤波,之后采用动态阈值法对图像进行二值化。细化采用的是模板比对法,将二值图中无关脊线方向的黑像素去除,达到细化的目的。所有上述方法都经Matlab验证,取得了较好的图像预处理效果。     

非接触指纹图像识别算法研究

非接触指纹识别具有接受程度高、防伪性高、卫生性高等优点,是目前生物特征识别领域的研究热点,但是非接触指纹图像的背景区域比接触式的相对复杂,指纹图像会出现旋转和平移现象,且脊、谷线的对比度较低,这些因素严重影响了非接触指纹的识别性能.采用接触式指纹图像处理方法很难达到良好的处理效果.本文根据非接触指纹图像的特点提出了相应的非接触指纹图像的预处理方法.首先采用图像YCbCr模型中的Cb分量和Otsu法相结合的方法进行手指区域的提取.其次采用高频强调滤波和迭代自适应直方图均衡化相结合的图像增强算法进行图像增强处理,再用简化的Gabor函数模板进行二次增强,然后提出了一种手指指纹ROI区域提取的方法.最后本文采用基于AR-LBP算法进行特征提取,利用最近邻分类器进行特征匹配.实验结果表明,本文提出的非接触指纹算法能够达到很好的图像识别效果.     

Real-Time Non-Driving Behavior Recognition Using Deep Learning-Assisted Triboelectric Sensors in Conditionally Automated Driving

Real-time recognition of non-driving behaviors is of great importance in conditionally automated driving, as it determines the takeover time budget, which in turn has a huge impact on the performance of the takeover. Here, a novel real-time non-driving behavior recognition system (RNBRS) integrating self-powered, low-cost, easy-to-manufacture triboelectric sensors, and a deep learning model is proposed. The structure, working mechanism, and electrical characteristics of triboelectric sensors are investigated and analyzed. Through the ingenious structural design of single-electrode triboelectric sensors and driving simulation experiments under conditional automated driving, non-driving behaviors are captured in the form of electrical signals. A well-trained long short-term memory network model is adopted to recognize the five most typical non-driving behaviors, including phone, console touchpad, driving, monitoring driving, and no operation, and test accuracy of 93.5% is achieved. Demonstration of a set of controlled experiments shows that RNBRS enables vehicles with conditional automation to dynamically adjust takeover time budget based on driver behavior, therefore significantly improving both safety and stability of takeover. This study opens new frontiers for the development of self-powered electronics and inspires new thoughts on human-machine interaction and the safety of autonomous vehicles.